Hi Amber Users,
We have been working on some MMPB(GB)SA calculations using the new implementation of AmberTools14 and we are getting strange results.
We have used two options for decomposing the total energy into components: idecomp=2 (per residue) or idecomp=4 (pairwise). They should not affect the total energy in the calculation. However, the total energies differ depending whether we use idecomp=2 or idecomp=4. Using GB, the results are the same for both idecomp options, but using PB the result differ.
Please, find below an example of this calculation on exactly the same system (only using idecomp=2 or 4).
FINAL RESULTS - Per Residue Decomposition (idecomp=2)
NSTEP ENERGY RMS GMAX NAME NUMBER
1 -2.0693E+03 1.8571E+01 1.2681E+02 C 1456
BOND = 348.5332 ANGLE = 949.0414 DIHED = 1280.5748
VDWAALS = -899.3273 EEL = -7849.1436 EPB = -1458.7532
1-4 VDW = 433.1028 1-4 EEL = 4767.9374 RESTRAINT = 0.0000
ECAVITY = 954.5187 EDISPER = -595.7736
FINAL RESULTS - Pairwise Decomposition (idecomp=4)
NSTEP ENERGY RMS GMAX NAME NUMBER
1 -6.2420E+02 1.8571E+01 1.2681E+02 C 1456
BOND = 348.5332 ANGLE = 949.0414 DIHED = 1280.5748
VDWAALS = -899.3273 EEL = -7849.1436 EPB = -13.6645
1-4 VDW = 433.1028 1-4 EEL = 4767.9374 RESTRAINT = 0.0000
ECAVITY = 954.5187 EDISPER = -595.7736
As it can be seen above, all the values are the same with the exception of the EPB (in red).
Could you please help us to clarify this issue?
Thanks very much!
Best,
Yoel
Mount Sinai School of Medicine, NYC
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue May 27 2014 - 10:30:02 PDT
